Create an Agent-lens application

Step 1: Enable Agent-lens service

1. Log on to the ApsaraDB for ClickHouse console.

2. On the Clusters page, select Enterprise Edition Clusters, find your target instance, and click its ID to open the instance details page.

3. In the left-side navigation pane of the instance details page, click Agent Observability Suite.

4. In the dialog box, click OK to enable the Agent-lens service.

Step 2: Configure IP whitelist

The first time you access the Langfuse service page, you must add your local IP address to the Langfuse service's IP address whitelist.

When the Instance Status changes to Running, check the IP address displayed in the dialog box. If it is your local IP address, click OK. If not, you must add your IP address to the IP address whitelist manually.

Step 3: Create user

1. Click the Public Address to go to the Langfuse service page.

2. The first time you access the page, click Sign up. Enter your name, email address, and password to create a Langfuse user.

Step 4: Create organization and project

You can select an existing organization at the top of the page or follow these steps to create a new one.

1. Click New Organization.

2. Enter an organization name and click Create.

3. Click Next.

4. Enter a project name and click Create.

Step 5: Create API key

You use a Langfuse API key to send observability data to Langfuse.

To create a new API key, click Create new API key.

Step 6: Send observability data

Langfuse seamlessly integrates with model providers such as Model Studio, OpenAI, and Anthropic. The following example uses Model Studio to demonstrate how to send your first observability data.

1. Install the Langfuse SDK.

   pip install langfuse

2. Set environment variables.

   Set the Langfuse API key and service endpoint as environment variables to direct the SDK's output to your Langfuse project. Set the Model Studio API key as an environment variable to allow the application to call the Model Studio model service.

   Note

   Before you send data, make sure the IP address of the SDK host is in the Langfuse service's IP address whitelist.

   export LANGFUSE_SECRET_KEY="sk-lf-..."
   export LANGFUSE_PUBLIC_KEY="pk-lf-..."
   export LANGFUSE_BASE_URL="http://{your_langfuse_endpoint}:3000"
   export DASHSCOPE_API_KEY="sk-..."

3. Create a file named quick_start.py and add the following code.

   import os
   from langfuse.openai import OpenAI
   
   try:
       client = OpenAI(
           api_key=os.getenv("DASHSCOPE_API_KEY"),
           base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
       )
   
       completion = client.chat.completions.create(
           model="qwen-plus",
           messages=[
               {'role': 'system', 'content': 'You are a helpful assistant.'},
               {'role': 'user', 'content': 'Who are you?'}
           ]
       )
       print(completion.choices[0].message.content)
   except Exception as e:
       print(f"Error: {e}")
       print("For more information, see the documentation: https://www.alibabacloud.com/help/en/model-studio/latest/error-codes")

4. Run the quick_start.py file to send the observability data to the Langfuse service.

   python quick_start.py

   Example response:

   Hello! I am Qwen, a large language model from Alibaba Group. I can answer questions, create text such as stories, official documents, emails, and scripts, perform logical reasoning, write code, express opinions, and play games. If you have any questions or need help, feel free to ask me anything!

   You can view the observability data you sent on the Langfuse service page: